Camera and Sun

Since I have started using V-Ray, something like 14 years ago, I have never been able to use a V-Ray Camera with a V-Ray Sun/Sky and real-world settings. Below is a photo to the left and my render to the right. I used the photo’s settings for the time of day and camera settings, but as you can see, nothing looks correct. I can go back to my normal workflow, but I want to know what I am doing wrong.

I think because you are looking at the raw render and the photo reference is tone-mapped .
also your walls are white -and maybe even too white (high albedo) which cause more light bouncing.
Also I heard that-I stand to be corrected - Vray camera is using different exposure table comparing to Max physical camera , so it worth checking with that too .
Those are just late night random thoughts, so I’ll wait for official comment on this : ))

I don’t thing the photo was tone mapped. My client took it yesterday and sent it from the job site.

I’m not expert but as far as i know every camera have a limited dynamic range and try to squeeze info into this range and apply some curve to make the image look “nice” but again this is just random thought on this as your issue could be caused by something else.

I think @^Lele^ can give -as always- better answer here : ))

But… open in Corona, convert, and render. I don’t have a problem there. Something inherently isn’t right in V-Ray. It might be me, which is why I am asking.

Well, as it was pointed out, it has to do with the shaders’ albedo, that i can tell.
That wood would be around 10% total reflectivity (putting a handy-wavy number to it!), while the scene you’re rendering is likely around 90% for those white walls.
The outside looks as the picture (ie. well burnt out, unlike for the HDRI), while on the inside you have a ton of bouncing, high-energy, low-decay light which can’t extinguish due to the high walls albedo.
You may also have to verify the sun angle, as if it’s even a bit off, it can change the sun’s intensities quite a bit.

Citing a conversion to Corona isn’t helpful (it only grows the number of variables at play), but a scene would definitely be. :smile:

I use SigarShaders, so maybe I rely too much on their numbers. I do know that their whites are more of a darker gray. For the sun, the photo has the time of day, location, and camera settings, so I am assuming that is accurate information.

So, the lighting is fine, the camera exposure is fine, what is left is the albedo of the shaders.
Make them *much* darker still (output map, or if present, output section of the bitmap), and it’ll start to get to the levels of light distribution you see in the photo.

I remember when you told us to make everything .255 darker. I am still mad at you for that :). I spent weeks updating all my assets and then weeks undoing it.

i remember a time when white was white, not grey :slight_smile:

See what i did to you.
Fifteen years later, and you’re still broken because of this.
I’m ever so sorry.

All this pent-up anger over the year. Grrr… :slight_smile:

BTW, I can easily correct things by tweaking the camera or sun intensity, however, that isn’t the point. I wonder if everyone plays with the camera exposure until it looks correct and the issue isn’t being addressed, or people stopped talking about it.

From my part, i matched live footage for years, without anything untoward from the side of V-Ray.
This is why i’m surprised by your difficulties, but then i didn’t have to match a thing in a long while, so you leave me with a modicum of doubt.
Could you maybe share just the basics of that shot, privately, with me?
Sun, the walls, the glass, or something along the lines?

We could continue the convo through email (so you could uncork the anger bottle and pour! ^^) and i could be more specific.

I’ve never experienced these things, The VraySun and the HDRIs I use for interior and exterior have always been in range of realistic camera exposure settings for a range of scenes. As Lele says its quite important to use the right albedo values and in my experience reflection as well. My whitest white is 180 and I tend to have good results with these. Photos like your example are almost always tonemapped and for an interior most likely even overexposed.

I need to stop being so busy and set aside some R&D time to dig deeper.

Provided it doesn’t lead to starving your family, make it a *recurring* priority.
It will save you (if it will work as it did for me…) countless hours of frustration and puzzlement.
But you won’t ever be able to do it all in one go, so you’ll need to set aside some time every so often, to keep abreast of changes in the market as a whole, and simultaneously dig deeper in the stuff which is most relevant to your work.
I won’t take responsibility for kids shrunk to 25.5% of their size, though! :smile:

I find absurd that there are still this kind of questions from users with years of experience behind.
These explanations have been given to us over and over again over the years.
If we succeed in getting good renders with old and wrong methods it does not prove that things have been done the right way, but only that we are able to make up for our technical deficiencies artistically, and that’s fine, until we complain.

On the other hand I believe this is a symptom of how little attention has been paid to inform users towards the correct use of tools in the past.
Most of us carry with a lot of errors of approach, shortcuts and settings of the software that instead of helping us, create fences from which we often fail to get out.​​

As far as I’m concerned I haven’t had any of these problems for years, since I started thinking about my work more as a virtual photographer.

- Use only real high quality hdri for lighting (or sun and sky).
- work on all your materials from scratch, using references, don’t use shaders collections ( follow lele’s suggestion about albedo vlaues)
- Re-check all the materials of the models you buy.
- Use realistic exposure values
- your render in the vfb without corrections has to look washed-out, to have a good preview of the final render always use a S curve correction o better use a good quality linear lut that simulates camera cc.

Maybe you’ll became less productive for a while but it’s worth it.

Exactly my two cents. About the productivity, you’ll lose some time in the beginning, but your renders will look beter and over time you’ll regain the time you lost because you won’t have to fiddle with wrong values anymore.

Hey, i need to justify my wage! ^^